write ahead logging
log-first, apply-late
The systems write the change to a WAL, instead of writing directly into the main storage. Its a way of to guarantee the ATOMIC principles inside the databases mechanisms.
Its very simple: All the changes are registered in a permanent storage log before its changes are applied in the main data store (as tables and indexes)
Is this related to CDC (change data capture) ?
Appending a log is way more efficient than making scattered random writes to a database. Storage systems are often optimize for sequential writes
WAL takes advantage of this by batching and appending changes, resulting in faster writes, lower IO overhead simplified recovery mechanisms
MongoDB uses a oplog (its WAL pattern)
walgit
References
https://www.architecture-weekly.com/p/the-write-ahead-log-a-foundation
https://www.postgresql.org/docs/current/wal-intro.html